Hp spectre x350 DA0X36MBAE0 no power

[crea2k]

Hi iv got an Hp spectre x360 DA0X36MBAE0 in for repair.
It was referred to me from another shop.

It was brought in as the other shop had told the customer it was dead because of a bad port. Although the port isn't great it's not shorted inside or mangled to the point of being unusable.

I have tried removing it anyway but makes no difference.

I looked at the motherboard and found that I had a short on the pp_hv 1 and 2 caps on the output of TPS65988 .

I diagnosed this as failed TPS65988 and removed it from the board.

The short went away, but even after replacing the ic there is no power.

Usb meter shows a pulse of 0.07mah every few seconds, but no other activity.

If I plug a power supply of 15.4v into the battery connector I get a amp draw of 0.15 on a similar time period to when on usb.

If I prove random rails (no idea what half of them are as no schemetic I get a power rail that appears for about 2 seconds then drops back to 0v, then repeats after 5 seconds or so.

I have probed a couple of coils and have found 3.3v and 1.05v with similar activity.

I have done a check of most of the large caps on the board and can't see any more shorts.

The behaviour would lead me to believe that they may be a short I'm not seeing, but the current draw is tiny, so I'm not sure.

[reformatt]

Re: Hp spectre x350 DA0X36MBAE0 no power

Make sure you have the exact same IC as what you took out, paying attention to the suffix. I don't have a schematic for it, but a quick check on Aliexpress shows there are multiple variants of this IC.

I ran into something similar with the earlier TPS65982DA on 15-W X360. I put in a plain TPS65982 and no port would negotiate to 20V with the charger. Wouldn't work till I put in the DA version. Get a USB C ammeter to confirm whether the port is being enabled via the CC pin, then check negotiated voltage with the charger.

[crea2k]

Re: Hp spectre x350 DA0X36MBAE0 no power

Yes iv got the tps65988ce which is marked on the ic as ptps65988ce G TI 78J P4L2
The original says tps65988ce TI 94J P0Y8 G4
This is the closest version I can find anywhere, so presume it works?.

I have a usb amp meter, it reads 0.07mah every few seconds.

Is there a specific supplier you suggest for the ic that works ?.

[reformatt]

Re: Hp spectre x350 DA0X36MBAE0 no power

Only thing you have to go off is the datasheet. How many of these IC's are there on the board? Using an USB C ammeter, are you getting 20V negotiated after plugging in? Check the LDO outputs of 1.8V and 3.3V too, maybe a short in that area?

According to the datasheet, PP_HV1 and PP_HV2 are system side, not port side. I'd have thought the port side VBUS would've sustained damage, as these connect to the outside world.

[crea2k]

Re: Hp spectre x350 DA0X36MBAE0 no power

There's one chip, it's dual channel for both ports.

So on the top of each coil I wrote the pulsed voltage. I then injected that voltage into each coil with the power supply without anything else plugged in.

All the coils but this large one near the battery connector with a mosfet on each side take the current just fine. This one though was clicking the power supply into over current / short every few seconds. I sprayed the board with freeze spray and couldn't see anything heating up. If I apply power when it's cold it's like a rapid fire on the power supply.

Weirdly it doesn't show as short on the multimeter.

If I connect my other power supply it takes about 0.1a, then a second later clicks to 3 amp short.

It's very odd.

Iv used diode mode from the drain to source on both mosfets and they read as a diode would, not shorted.

[reformatt]

Re: Hp spectre x350 DA0X36MBAE0 no power

Voltage injection can be dangerous, especially if you don't have a schematic. Shorted buck converter FET's can result in a low voltage IC like a PCH or CPU copping the full rail voltage, so you should only use 1V during injection. What seems to save some devices is the external power adaptor immediately goes into protection mode in that event.

With 1V, this is difficult to trace without a thermal camera. I honestly recommend getting one of these (I have a Seek pro) as the time it saves you will pay itself off in no time.

Measurements alone on a rail aren't a definitive measurement, as the short could be after an active device. 1V and thermal camera are the way forward without a schematic.

[crea2k]

Re: Hp spectre x350 DA0X36MBAE0 no power

So it's working now as far as charging the battery goes. What I'm getting now is white light on the charger turning on for 5 seconds, then off for 7 seconds. It does this with the battery or the charger plugged in. The battery voltage is about 14.9-15.2v . It was 13v originally but left it charging overnight.

I've got a thermal camera on order from China at the moment. It's the quanli super cam X.

[crea2k]

Re: Hp spectre x350 DA0X36MBAE0 no power

I've noticed if I unplug the battery and then plug it back in with no charger plugged in it does this flashing cycle then, even with no prompt to boot.

[reformatt]

Re: Hp spectre x350 DA0X36MBAE0 no power

HP's often use the charging light to report POST errors. Is it blinking in a sequence?

[crea2k]

Re: Hp spectre x350 DA0X36MBAE0 no power

Yes it's 7 seconds off, 5 turned on.

[reformatt]

Re: Hp spectre x350 DA0X36MBAE0 no power

Just had a 13-AP0000tu come in with a DA0X36MBAE0 motherboard. Fault presentation was not charging. The rear USB-C port on the corner, only +5V negotiated when cable plugged in both ways. The other side mounted one, dead in one orientation, 20V negotiated on the other but cutting out if cable was disturbed. It did charge, but I noted current would go up to 3A, then back down to 200ma in a cycle.

Removed the board, there are two AON7403 FETs shorted, visible since they had melted the plastic PCB cover. They were fused to the board and couldn't be removed. Board had begun to carbonise so likely I'd end up with a short between layers if I went any further.

The one working USB-C port was flakey, and probably needed a replacement TPS65988CE IC. There's another controller in there that appears to connect to the USB ports as well, but cannot locate what that is. Deemed a no fix as multiple faults. Images attached as reference as no schematics.

HP make some good lappies, this is not one of them.

[mcplslg123]

Re: Hp spectre x350 DA0X36MBAE0 no power

Applying a lot of bga flux and solder on mosfet and holding 2 knife edged tip solder iron on both sides will perhaps remove them. Or you can break the top part of mosfets with surgical knife and then hold the knife edge solder iron on them. They will come out easily.

[reformatt]

Re: Hp spectre x350 DA0X36MBAE0 no power

Yes true, however have no boardviews or schematics if things go awry. Would also have to replace at least one USB-C port and run jumpers for ripped pads too. Effort vs cost customer is willing to pay is a factor so decided not to go ahead with it.